Genetically modified adenoviruses that replicate only in cancer cells containing certain mutation have shown promise as a new treatment for cancer. The adenovirus mutant ONYX-015 in particular has shown some clinical anti-tumor activity, although to a lesser extent than expected. Improvements of this promising therapeutic approach are therefore needed and can be achieved through a better understanding of viral and host factors determining therapeutic efficacy. A crucial step in viral life cycle is successful entry of the viral particle into target cells, which is strongly dependent on the presence of the main receptor for adenovirus, the coxsackievirus and adenovirus receptor (CAR). This receptor is frequently down-regulated in highly malignant cells, rendering this population less vulnerable to viral attack. We have shown that disruption of signaling through the RAF-MEK-ERK pathway by inhibition of MEK can up-regulate CAR expression, resulting in enhanced adenovirus entry into the cells. This pharmacological intervention, however, can also negatively interfere with the replication of ONYX-015 through its effects on the cell cycle. We hypothesize that efficacy of this treatment can be increased by optimizing the overall rate of virus spread: that is, the viral infectivity needs to be increased via the up-regulation of CAR, while maintaining the ability of the virus to replicate in cells. We will achieve this by using a combination of computational and experimental approaches that will be applied to cell line models of pancreatic cancer. Computational models provide an essential tool to search for optimal treatment regimes. We will use ordinary differential equations, based on previous work, to describe the dynamics between replicating viruses and growing tumors;in addition, we will employ control theory (commonly used in engineering) to search for optimal treatment regimes. Computational modeling will be closely tied to experiments, which will involve measurements of crucial parameters, discovery of potential new factors influencing adenovirus infection of cancer cells, model validation, and tests of predictions. This research will be beneficial for public health through improving cancer treatments that make use of viruses that are capable of killing cancer cells. The proposed studies will be performed using pancreatic cancer as a model disease. However, the results of these studies will be applicable to other types of cancer, thus potentially contribute to improving the treatment outcome for many cancer patients in the future.